JP2015014864A - Guidance system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To allow a user to safely move by providing a guidance route on which it is difficult for a suspicious person to find the user.SOLUTION: In a guidance system configured of: a detection part S for detecting a person; a portable terminal 2 owned by a user; and a guidance device 1, the guidance device 1: detects the position of the person by using a preliminarily stored space map and the output of the detection part S; determines whether the person is a suspicious person or the user; searches a coverage until light radially projected from the position of the suspicious person in a direction of a predetermined range is blocked by a building structure as a visual field model; evaluates a route from the position of the user to a doorway D by using the visual field model; and transmits a route on which it is difficult for the suspicious person to find the user to the portable terminal 2 as a guidance route.

Description

  The present invention relates to a guidance system that safely guides students and people working in the office when a suspicious person enters the school or office.

  Conventionally, the movement path of a suspicious person is identified using detection output from a personal sensor, the position of a user (children, etc.) is detected by GPS or the like, and the user is detected using the movement path of the suspicious person and the position of the user. A system that allows a user to evacuate safely from a suspicious person who has entered the house by calculating the guidance route (evacuation route) and displaying the guidance route as visible information on a portable terminal possessed by the user Has been proposed (Patent Document 1).

JP 2003-288657 A

  However, in the above prior art, although the guide route is calculated in consideration of the relative position between the suspicious person and the user and the movement route of the suspicious person, the viewpoint of easiness to find from the suspicious person is taken into consideration. Absent. Therefore, no matter how far away from the suspicious person, once they are discovered by the suspicious person, threats and threats from the suspicious person will threaten children and women, making it difficult to evacuate properly. obtain.

  Therefore, the present invention considers the suspicious person's visual field range (visible range) from the building structure and the suspicious person's position, evaluates the user's easiness of finding from the suspicious person, and uses a difficult-to-find guidance route. The purpose is to guide the person to make safer blame by making it possible to present to the person.

In order to solve such a problem, a detection unit installed in a monitoring space for detecting a person, a portable terminal held by the user, and a guide route that is difficult to find from a suspicious person from the current position of the user to the target position A guidance system comprising a guidance device that calculates and transmits the guidance route to the portable terminal,
The guidance device includes a space map indicating a structure of the monitoring space and an installation position of the detection unit, a storage unit that stores in advance a target position on the space map, an output of the detection unit, and the space map. A person detecting means for detecting the position of the person existing in the monitoring space using a person, a person determining means for determining whether the person existing in the position of the person is a suspicious person or a user, and a suspicious person position in the space map One or a plurality of guiding paths from the user position to the target position using the visual field model by determining the reach range until the light projected radially in the direction of the predetermined range is blocked by the structure And a guide route generating means for evaluating each candidate and generating the guide route candidate that is most difficult to find from the suspicious person as the guide route. To provide.
With this configuration, the guidance system of the present invention calculates the visual field range of the suspicious person as the visual field model, calculates the guidance route that is most difficult to find from the suspicious person using the visual field model, and possesses the guidance route by the user. Since it can transmit to a portable terminal, the user can move to a target position more safely.

Further, according to a preferred aspect of the present invention, the guidance route generation means estimates the front direction of the suspicious person using the output of the detection unit, and the guidance route candidate located in the front direction is more likely to be found. Shall be evaluated.
With this configuration, the guide route candidate that must move the position of the suspicious person in the front direction is evaluated as being easily found by the suspicious person. Therefore, the user moves along the guide route that is not located in the front direction as much as possible. Can be moved to the target position more safely.

In a preferred aspect of the present invention, the guidance route generation means estimates the suspicious person position in an evaluation period after the current time using a time transition of the output of the detection unit, and the suspicious person position and the spatial map Is used to estimate the visual field model in the evaluation period, the user position in the evaluation period is estimated for each guidance route candidate, and evaluation is performed for each guidance route candidate using the visual field model and the user position. The ease of finding from a suspicious person in a period is evaluated, and the guidance route candidate that is most difficult to find from a suspicious person is generated as the guidance route.
With this configuration, the position of the suspicious person is estimated by estimating the position of the suspicious person at the future time and evaluating the easiness to find from the suspicious person using the visual field model at the position of the suspicious person and the user position at the future time. It is possible to obtain a guidance route that is difficult to find in consideration of the above, and to provide information on the guidance route with higher safety to the user.

In a preferred aspect of the present invention, the structure in the spatial map is stored as a three-dimensional shape, and the visual field model is a horizontal position of a suspicious person in the spatial map and a height position corresponding to an eye position. It is assumed that it has a three-dimensional shape consisting of an arrival range until light projected radially from the position in a predetermined direction is blocked by the structure.
With this configuration, the hiding of the suspicious person's line of sight due to the structure can be calculated more accurately, and the visual field range of the suspicious person can be more faithfully reproduced.

In a preferred aspect of the present invention, the guidance route generation means calculates a separation distance between the suspicious person position and the user position, and calculates a total distance of a route from the user position to the target position. Assume that the smaller the separation distance is, the easier it is to be found from a suspicious person, and the larger the total distance is, the easier it is to be found from a suspicious person.
With this configuration, a guide route candidate with a larger separation distance is more likely to be set as a guide route, and a guide route candidate with a larger total distance is less likely to be set as a guide route. Therefore, the user can be presented with a guide route that can reach the target position as soon as possible and can be as far away as possible from the suspicious person, so that a safe guide route that is difficult to find comprehensively can be provided to the user.

  As described above, the guidance system of the present invention can present to the user a guidance route that is difficult to find from the suspicious person by considering the range that can be seen by the suspicious person.

Overview of guidance system The figure explaining the monitoring range of a detection part Block diagram of guidance device Diagram explaining the visual field model The figure showing the example of a detection signal table Diagram showing an example of a tracking table Flowchart of control unit processing Flow chart of guidance route generation processing Diagram showing examples of guide route candidates Diagram explaining the calculation of awareness Flowchart of guidance route generation processing in the second embodiment Diagram explaining suspicious person position at future time Diagram explaining user position at future time The figure explaining calculation of the degree of detection in 2nd embodiment Diagram explaining total distance and separation distance A diagram for explaining an example of a weighted visual field model Example of 3D spatial map and visual field model

(First embodiment)
Hereinafter, as one embodiment of the present invention, when a building is set as a monitoring space, and a suspicious person enters the monitoring space, a destination position that is a guidance destination in an evacuation route (guidance route) that is difficult to be found by the intruder ( A guidance system for guiding to the doorway will be described with reference to the drawings.

  FIG. 1 shows an overview of the guidance system used in this embodiment. As shown in FIG. 1, the building B forming the monitoring space in the present embodiment includes a passage P painted by hatching, an entrance D, and windows indicated by reference numerals W1 to W9. Moreover, since the transparent glass material is used for the window W, the person who exists in the passage P can see the courtyard of the building B through the windows W4 to W7, and the windows W1 to W3, W8, and W9. It is possible to see the outside of the building B through The guidance system includes a detection unit S, a guidance device 1, a portable terminal 2, a network 3, and a wireless communication device 4.

  On the ceiling portion in the passage P of the building B, a detection unit S indicated by reference numerals S1 to S29 is installed. The detection unit S is an infrared sensor that receives infrared rays emitted from the human body of a person located within the detection range and detects the presence or absence of a human body or the like based on a change in the amount of received light. In the present embodiment, as shown in FIG. 2, the detection ranges of the detection units S1 to S29 are set as ranges indicated by hatching within the dotted lines of reference signs A1 to A29, respectively. The detection units S1 to S29 are connected to the network 3 including a LAN line or the like via a communication line. When a person is detected within the detection range A, the detection unit S1 to S29 records the identification number and detection time of the detected detection unit S. The detected signal is transmitted to the guidance device 1 via the network 3.

  The guidance device 1 is a computer installed in an office room, a security room, a server room (not shown) in the building B, and the user's safety based on the detection signal of the detection unit S transmitted via the network 3. A process of calculating the guide route and transmitting the calculated guide route information to the mobile terminal 2 possessed by the user is performed. At this time, the guidance device 1 transmits information on the guidance route calculated by wireless communication to the mobile terminal 2 via the wireless communication device 4 connected to the network 3. The wireless communication device 4 is a communication device such as a wireless LAN router or a wireless access point.

  The portable terminal 2 is an electronic device having a display screen 21 such as a smartphone, a PDA, or a tablet terminal, and is carried by a user Y who is a regular user in a building B such as a child. The mobile terminal 2 receives the information on the guidance route transmitted from the guidance device 1 via the wireless communication device 4, and based on the information on the guidance route, displays image information indicating the guidance route on the map of the building B. A process of generating and displaying on the display screen 21 is performed. The user Y can grasp a safe guide route that is difficult for a suspicious person to find by referring to the guide route displayed on the display screen 21.

FIG. 3 shows the configuration of the guidance device 1. In order to realize the functions described above, the guidance device 1 of the present embodiment is schematically configured to include a storage unit 11, a control unit 12, an input unit 13, and a communication unit 14.
The input unit 13 is an information input device such as a keyboard, a mouse, a touch panel, or a portable storage medium reading device. A system administrator or the like uses the input unit 13 to set various types of setting information such as a space map 111 (to be described later) that is a map of the building B with respect to the guidance device 1, installation position information of the detection unit S on the map, Can be entered.

  The storage unit 11 is an information storage device such as a ROM (Read Only Memory), a RAM (Random Access Memory), and an HDD (Hard Disk Drive). The storage unit 11 stores various programs and various data, and inputs / outputs such information to / from the control unit 12. The various data includes a spatial map 111, a target position 112, a visual field model 113, a detection signal table 114, a tracking table 115, various parameters 116 such as setting values used for each process of the control unit 12, and the control unit 12 Calculation results and the like of the guidance route and the like finally calculated obtained by the processing are included.

  The space map 111 is so-called map information indicating the structure of the building B that is the monitoring space and the installation position of the detection unit S, and is stored in the storage unit 11 as coordinate information. In addition, an attribute is given to each component constituting the structure of the building B. For example, an attribute indicating the type of a component such as a window W or an entrance / exit D, or an attribute indicating a material of the component such as the window W being a light-transmitting material is given. The space map 111 is stored in the storage unit 11 by being registered from the input unit 13 by an administrator or the like.

  The target position 112 is position information in the spatial map 111 indicating a place where the user Y should be guided, and is stored in the storage unit 11 as coordinate information. In the present embodiment, the coordinate information of the gravity center position of the entrance D of the building is set in advance by an administrator or the like with the entrance D where the user Y can ensure safety as the target position 112. However, the present invention is not limited to this, and the embodiment may be such that the target position 112 is set as the position of a child who is late to escape. As a result, when a child who is late to escape is rescued, a rescue route (guide route) that is difficult to be found by a suspicious person can be presented to a user (for example, a rescuer such as a teacher). Furthermore, an embodiment in which a plurality of target positions 112 are set in advance and the target positions 112 to be employed are automatically selected according to the situation may be used. For example, the destination position 112 closest to the current user position may be automatically selected from a plurality of destination positions 112 registered in advance.

  The visual field model 113 is information obtained by modeling the visual field range of the suspicious person X (a range that can be almost seen) in association with the spatial map 111, and is stored in the storage unit 11 as coordinate information in the spatial map 111. The visual field model 113 is appropriately generated by the guidance route generation means 123 of the control unit 12 to be described later using the position information of the suspicious person X (hereinafter referred to as “suspicious person position”) and the space map 111, and is stored in the storage unit 11. Remembered. FIG. 4 is a diagram for explaining the visual field model 113. FIG. 4A shows the visual field model 113 of the suspicious person X at the time t0, and FIG. 4B shows the suspicious person X at the time t1. The visual field model 113 is shown. As shown in FIG. 4, the visual field model 113 is configured such that light projected radially in a predetermined range direction (for example, a range direction around 360 °) centered on the position of the suspicious person is blocked by the structure (wall surface, etc.) of the building B. The reachable range until it is obtained is obtained as the visual field model 113. In the present embodiment, the 360 ° range is used as the “predetermined range direction” for projecting light. However, the present invention is not limited to this, and the traveling direction of the suspicious person X is 0 ° and ± 70 °. The direction of the range may be adopted as the direction of the predetermined range. In FIG. 4, the light reaching range M is set up to a preset radius L [m] (for example, 20 [m]). However, the present invention is not limited to this, and the light reaching range M is not set ( The visual field model 113 may be calculated by setting (∞).

  The detection signal table 114 is a table created and updated based on the detection signal received from the detection unit S. As shown in FIG. 5, the identification number of the detection unit S that has transmitted the received detection signal, and It is a table which matches the detected time and the person position calculated by the person detection means 121 mentioned later based on the detection signal. In the present embodiment, the coordinates of the installation position of the detection unit S corresponding to the received identification number of the detection unit S are used as the person position. However, the present invention is not limited to this, and the coordinates of the gravity center position of the detection range A of the detection unit S may be set as the person position.

  The tracking table 115 is a table that is created and updated based on information such as the detection signal table 114, and as shown in FIG. 6, the time and the identification number of the person tracked by the person detecting means 121 described later. (Person ID), a person position of the person at the time, and a person attribute (suspicious person / user) determined by the person determination unit 122 described later.

  The communication unit 14 is a communication interface such as a LAN or USB, and communicates with the detection unit S, the mobile terminal 2, and the wireless communication device 4 via the network 3.

  The control unit 12 is an arithmetic device such as a CPU or a DSP, for example, and executes various types of information processing according to programs stored in the storage unit 11. In the present embodiment, the control unit 12 receives a detection signal from the detection unit S and performs a series of processes shown in FIGS. 7 and 8 to be described later. The route generation means 123 is included. In addition, the control unit 12 performs processing for storing input information such as setting information and operation information from the input unit 13 in the storage unit 11.

  The person detection unit 121 detects a position of a person existing in the monitoring space based on a detection signal from the detection unit S received via the network 3 and tracking for tracking the detected person in time. Process. In the position detection process, the detection signal table 114 is updated based on the received detection signal. For example, assume that detection signals are received from the detection units S2 and S22 when the current time is t0 as in the example of FIG. In this case, it is determined that there is a person in a position within the detection ranges A2 and A22 corresponding to these detection units S2 and S22, and the detection signal table 114 is updated to add the position together with the detection time and the detection signal. Process as follows. Next, in the tracking process, based on the positional relationship of the person positions from the past time to the current time in the detection signal table 114, the process of tracking the person in time by associating the same person at each time is performed. For example, in the example of FIG. 6, when a person is detected at the position of the detection range A2 at time t0, a detection range that is a position adjacent to the position A2 without detecting a person at the position A2 at the next time t1. When a person is detected at A3, assuming that the person has moved from the position A2 to the position A3, these persons at the previous and subsequent times are associated as the same person. That is, the tracking table 115 at the current time t1 is updated so that the same person ID (= X) as the person ID (= X) existing at A2 at the previous time t0 is given.

  The person determination unit 122 performs a person determination process for determining whether the person first detected by the person detection unit 121 is a suspicious person or a user. In this embodiment, on the assumption that the entrance / exit D is appropriately controlled, the person entering the entrance / exit D is determined to be an appropriate person (user), and the inside of the building B from a position other than the entrance / exit D It is determined that the person who has intruded is an inappropriate person (suspicious person). That is, when a person is detected for the first time in a detection range other than the detection range A10 where the entrance D is located, the person is determined to be a suspicious person. For example, when the position at which a certain person is detected for the first time is A1, it is assumed that the person is a suspicious person who has entered through the window W1, and the attribute corresponding to the person ID of the person in the tracking table 115 is “suspicious person”. Update. On the other hand, when the position at which a certain person is detected for the first time is A10, it is assumed that the person is an appropriate user who has entered through appropriate entrance / exit management (for example, authentication at the time of admission). The attribute corresponding to is updated as “user”. Note that a person whose attribute has been determined is not subject to the person determination process in the subsequent processes, and the same person being tracked is updated to have the same attribute.

  The guide route generation unit 123 performs a guide route generation process for generating a safe user guide route that is most difficult to find from a suspicious person. Details of the guidance route generation process will be described later.

(Regarding the processing operation of the control unit 12 of the guidance device 1)
Next, a processing operation of the control unit 12 until a guidance route is generated using the guidance device 1 having the above-described configuration and the guidance route is transmitted to the mobile terminal 2 possessed by the user will be described with reference to FIG.

  When the guidance device 1 is activated, the control unit 12 executes detection signal acquisition processing (ST1). In the detection signal acquisition process, a detection signal transmitted when the detection unit S detects a person is received, and the detection signal table 114 is associated with the detection time described in the detection signal and the identification number of the detection unit S. Perform the update process.

  Next, the person detection means 121 of the control unit 12 performs the identification based on the identification number of the detection signal received in ST1 and the spatial map 111 (in which the installation position of the detection unit S is written) of the storage unit 11. A position detection process for detecting a person position corresponding to the number is executed (ST2). That is, a process of obtaining the installation position of the detection unit S corresponding to the received identification number from the space map 111, detecting the coordinates of the installation position of the detection unit S as a person position, and storing the person position in the detection signal table 114 I do.

  Next, the person detection unit 121 of the control unit 12 associates persons between the times based on the positional relationship of the person positions from the past time to the current time in the detection signal table 114 updated in ST1 and ST2. A tracking process for tracking a person in time is performed (ST3). Since the details of the tracking process have been described above, the description thereof is omitted here.

  Next, based on the tracking table 115, the control unit 12 determines whether a person exists in the monitoring space at the current time (ST4). Specifically, when at least one person with a person ID is stored in the tracking table 115 at the current time, it is determined that there is a person in the monitoring space (ST4-Yes), and ST5 Proceed with the process. On the other hand, if at least one person with a person ID is not stored in the tracking table 115 at the current time, it is determined that no person exists in the monitoring space (ST4-No), and the process proceeds to ST1. .

  Next, the person determination unit 122 of the control unit 12 determines whether the person detected for the first time is a suspicious person or a user in the processes from ST1 to ST3, and updates the attribute of the tracking table 115. A determination process is performed (ST5). In the present embodiment, a person for whom attribute information in the tracking table 115 is not stored is a person who is detected for the first time, and the person is a target person for person determination processing. Then, when the current position of the target person is the position of A10, it is determined that the target person is a user, and when it is a position other than A10, the target person is determined to be a suspicious person.

  Next, the control unit 12 refers to the tracking table 115 of the storage unit 11, and whether or not “the suspicious person exists and the user exists” in the monitoring space at the current time is satisfied. Is determined (ST6). At the current time, when “a suspicious person exists and a user also exists” (ST6-Yes), the control unit 12 advances the process to ST7. On the other hand, when “the suspicious person does not exist” or “the user does not exist” at the current time (ST6-No), the control unit 12 advances the process to ST1.

  Next, the guide route generation means 123 of the control unit 12 performs guide route generation processing for generating a safe user guide route that is most difficult to find from a suspicious person existing in the monitoring space (ST7). Details of the guide route generation process will be described later.

  Next, the control part 12 performs the process which transmits the guidance route produced | generated in ST7 to the portable terminal 2 which a user possesses (ST8). In the present embodiment, image information obtained by combining the guidance route generated in ST7 and the space map 111 is transmitted to the mobile terminal 2. Then, the mobile terminal 2 that has received the image information displays the image information on the display screen 21. However, the present invention is not limited to this, and the coordinate information of the guidance route generated in ST7 is transmitted to the portable terminal 2, and the portable terminal is stored using the spatial map 111 stored in advance on the portable terminal 2 side and the received coordinate information of the guidance route. Image information obtained by combining the guidance route and the space map 111 may be generated on the second side.

(Details of guidance route generation processing)
Next, details of the guidance route generation processing executed by the guidance route generation means 123 of the control unit 12 in ST7 of the flowchart of FIG. 7 will be described with reference to FIG.

  First, the guidance route generation unit 123 of the control unit 12 reads the current position (suspicious person position) of the suspicious person X from the tracking table 115 of the storage unit 11, and uses the suspicious person position and the spatial map 111 of the storage unit 11. Then, a visual field model calculation process for calculating the visual field model 113 is performed (ST11). As described above, the guide route generation unit 123 has the structure of the building B in which the light projected on the visual field model 113 in a predetermined range direction (in this embodiment, a range direction around 360 °) centered on the position of the suspicious person. It is calculated as the reach until it is blocked by.

  Next, the guidance route generation means 123 performs guidance route candidate calculation processing (ST12). In the guidance route candidate calculation process, the current position (user position) of the user Y is read from the tracking table 115 of the storage unit 11, and the user guidance is performed using the user position and the space map 111 of the storage unit 11. A “guide route candidate” that is a route candidate is calculated. In this embodiment, all the routes from the user position to the target position 112 are obtained as guide route candidates. 9A and 9B are diagrams showing examples of guidance route candidates. FIG. 9A shows a guidance route candidate R at time t0, and FIG. 9B shows a guidance route candidate R at time t1. As shown in the figure, at the user position of the user Y at time t0, three routes R10 to R30 represented by broken lines are obtained as the guide route candidates R, and the user position of the user Y at time t1. Then, three routes R11 to R31 are obtained as the guide route candidate R.

  Next, the guidance route generation unit 123 calculates the degree of awareness D indicating an index of ease of finding from a suspicious person when the user moves the guidance route candidate for each guidance route candidate R calculated in ST12. An awareness level calculation process is performed (ST13). In the degree of awareness calculation process, using the visual field model 113 calculated in ST11 and the guidance route candidate R calculated in ST12, the degree of awareness D is calculated from the distance value of the guidance route candidate R superimposed on the visual field model 113. Is calculated. That is, the degree of awareness D is calculated larger as the guide route candidate R is more likely to be included in the range of the visual field of the suspicious person. FIG. 10 is a diagram for explaining the calculation of the degree of awareness D, where (a) illustrates the degree of awareness D at time t0, and (b) illustrates the calculation of the degree of awareness D at time t1. In the same figure, the part shown with the thick continuous line shown on the guidance route candidate R shown with a broken line is the place which the visual field model 113 and the guidance route candidate R have overlapped, and the distance value of the said superimposed part is large. As it is, the degree of awareness D is greatly calculated. For example, in FIG. 9A, the degree of awareness of the guidance route candidate R10 = 0 [m], the degree of awareness of the guidance route candidate R20 = 5 [m], and the degree of awareness of the guidance route candidate R30 = 30 [m]. Is done. In FIG. 5B, the degree of awareness of the guidance route candidate R11 = 0 [m], the degree of awareness of the guidance route candidate R21 = 10 [m], and the degree of awareness of the guidance route candidate R31 = 40 [m]. Is done.

  Next, the guide route generation means 123 performs a process of setting a guide route candidate having the smallest degree of awareness D as a guide route to be transmitted to the user from the guide route candidates R calculated in ST13 (ST14). . In the example of FIG. 10, the guide route candidate R10 is set as the guide route when the current time is t0, and the guide route candidate R11 is set as the guide route when the current time is t1.

  As described above, since the guidance system according to the present embodiment can present a guidance route that is difficult to be found by a suspicious person to the user, the user can safely move so that the suspicious person cannot find it.

(Second embodiment)
In the first embodiment described above, the degree of awareness D is calculated based on the position of the suspicious person at the current time without considering the movement of the suspicious person X at the future time, and the guidance route is set only from the degree of suspicion D. ing. However, since it is assumed that the suspicious person X is actually moving, even if the degree of awareness D is the smallest at the current time, the degree of awareness D of the guidance path is the smallest at the future time. Not exclusively. Even if the degree of awareness D is small, the distance to reach the target position 112 is too large, or it is difficult to find a guidance path that passes near the suspicious person X. Is not limited. Therefore, in consideration of the position of the suspicious person at the future time, the distance value of the guide route candidate R (hereinafter referred to as “total distance”) and the distance from the suspicious person X are considered to be the safest overall. An embodiment for the purpose of presenting a guidance route to a user will be described.

  The overall configuration of the present embodiment and the functional configuration (block diagram) of the guidance device 1 are the same as those of the first embodiment described above, and are represented in the same manner as in FIGS. In addition, description of each component included in the drawings is omitted. Moreover, since the flowchart showing the process of the control part 12 of this embodiment is also represented by FIG. 7 like 1st embodiment, description of the process is abbreviate | omitted here. Below, the same code | symbol is attached | subjected and demonstrated about the part similar to said 1st embodiment among the guidance apparatuses 1 of this embodiment. Since the difference between the present embodiment and the first embodiment is the guidance route generation process, the present embodiment will be described below using the flowchart of FIG. 11 representing the guidance route generation process of the present embodiment. In the flowchart of FIG. 11, loop 1 means that each process of ST24 to ST28 is executed by the number of guidance route candidates R calculated in ST23, and loop 2 evaluates each process of ST25 to ST27. This means that only the number of times is executed. Here, the evaluation time is a time when an evaluation period from the present to a predetermined future time is divided by a predetermined unit time interval, and is a time for calculating an instantaneous evaluation value to be described later. In the present embodiment, as the evaluation time, the evaluation time is set to 30 seconds after every 10 seconds from the current time. That is, when the current time is t = 0 [seconds], four evaluation times are set as evaluation times: 0 [seconds], 10 [seconds later], 20 [seconds later], and 30 [seconds later]. Yes. Therefore, in loop 2, the processes of ST25 to ST27 are repeated for the four evaluation times. The evaluation time can be set as an arbitrary value.

  In the guide route generation process of the present embodiment, first, a suspicious person position estimation process for estimating a suspicious person position at each evaluation time is performed (ST21). In the suspicious person position estimation process, first, based on the past movement history of the suspicious person in the tracking table 115 and the space map 111, the moving speed and moving direction of the suspicious person are estimated. Then, using the suspicious person position at the current time and the estimated moving direction, the suspicious person's movement path at the future time of the suspicious person (hereinafter referred to as “suspicious person expected path”) is estimated. At this time, in this embodiment, the behavior pattern of the suspicious person is stored in the storage unit 11 in advance, and the guidance route that is most likely to move from the suspicious person position and the estimated moving direction is selected as the suspicious person expected route. To do. Then, the suspicious person position at each evaluation time is estimated using the estimated suspicious person expected route and the estimated moving speed of the suspicious person. FIG. 12 is a diagram for explaining the suspicious person position at the future time estimated in the suspicious person position estimating process. In the figure, the route indicated by the symbol r is a suspicious person expected route, and the positions indicated by the symbols X0 to X3 are suspicious person positions at the respective evaluation times.

  Next, a visual field model calculation process for calculating the visual field model 113 at each evaluation time is performed using the suspicious person position estimated in ST21 and the space map 111 (ST22). Subsequently, a guidance route candidate generation process for generating a guidance route candidate using the user position and the space map 111 is performed (ST23). Since the visual field model calculation process and the guidance route candidate generation process here are the same processes as the visual field model calculation process (ST11) and the guidance route candidate generation process (ST12) of the first embodiment, detailed description will be given here. Is omitted.

  Next, user position estimation processing for estimating the user position at each evaluation time is performed (ST24). In the user position estimation process, the user position at each evaluation time, which is a future time, is estimated using the guide route candidate R currently processed in loop 1 and a preset standard moving speed. FIG. 13 is a diagram for explaining the user position at the future time estimated by the user position estimation process. In the figure, the positions indicated by symbols Y0 and Y11 to Y13 are user positions at the respective evaluation times in the guide route candidate R10. The positions indicated by the symbols Y0 and Y21 to Y23 are the user positions at each evaluation time in the guidance route candidate R20. The positions indicated by reference signs X0 and Y31 to Y33 are user positions at the respective evaluation times in the guide route candidate R30.

  Next, at the user position at the evaluation time that is the processing target in loop 2, the degree-of-detection calculation process is performed to calculate the degree-of-detection D that indicates an index of ease of finding from a suspicious person (ST25). In the present embodiment, it is determined whether or not the user position obtained in ST24 is included in the range of the visual field model 113 obtained in ST22. When included, “1” is not included. Finds the degree of awareness D as “0”. FIG. 14 shows (a) degree of awareness at evaluation time t = 0 (user position Y0) and (b) evaluation time t = 10 (user position) when the guide route candidate R to be processed is R10. It is a figure explaining calculation of the awareness degree D in Y11). In the example of FIG. 9A, the user position Y0 is not included in the range of the visual field model 113 at the evaluation time t = 0 (the visual field model 113 at the suspicious person position X0), and thus the calculated awareness D = 0 It becomes. Further, in the example of FIG. 5B, since the user position Y11 is included in the range of the visual field model 113 (the visual field model 113 at the suspicious person position X1) at the evaluation time t = 10, the calculated awareness D = 1

  Next, from the user position at the evaluation time to the target position 112 using the guide route candidate R that is the processing target in the loop 1 and the user position at the evaluation time that is the evaluation target in the loop 2. The total distance L1 is calculated. Subsequently, a separation distance L2 between the user and the suspicious person is obtained from the user position and the suspicious person position at the evaluation time. In the present embodiment, one or a plurality of routes from the user position to the suspicious person location are obtained, and the distance of the shortest route among the routes is obtained as the separation distance L2. FIG. 15 is a diagram illustrating an example of the total distance L1 and the separation distance L2 when the guide route candidate to be processed is R1 and the evaluation time t = 0. Respective values are calculated from the length of the total distance L1 indicated by the solid line and the length of the separation distance L2 indicated by the broken line.

  Next, using the degree of awareness D obtained in ST25 and the total distance L1 and the separation distance L2 obtained in ST26, an instantaneous evaluation value e that is an index representing the easiness to find from the suspicious person X at the evaluation time is calculated ( ST27). In the present embodiment, the instantaneous evaluation value e = α × arousal degree D + β × total distance L1−γ × separation distance L2. Here, α, β, and γ are preset in the storage unit 11 as positive coefficients. Which value among the degree of awareness D, the total distance L1, and the separation distance L2 is used to evaluate safety. It is a value that can be adjusted by the administrator in consideration of whether or not it is particularly important. As described above, the instantaneous evaluation value e is calculated to be larger as the awareness degree D and the total distance L1 are larger, and is calculated to be smaller as the separation distance L2 is larger.

  Next, a process for calculating a route evaluation value representing the final risk of the guidance route candidate to be processed in loop 1 is performed (ST28). In the present embodiment, the route evaluation value is calculated from the sum of the instantaneous evaluation values at each evaluation time calculated by each process of the loop 2.

  Next, using the route evaluation value of each guide route candidate R calculated in loop 1, a process of setting the guide route candidate R having the minimum route evaluation value as a guide route to be transmitted to the user is performed (ST29). ).

Thus, in this embodiment, a suspicious person position is estimated by estimating the suspicious person position in the future time, and calculating the awareness D using the visual field model 113 in the suspicious person position and the user position in the future time. Since the guide route that is difficult to find is calculated in consideration of the movement at the future time, information on the guide route with higher safety can be provided to the user. Further, in the present embodiment, when calculating the guide route, the total distance L1 that is the distance from the user position to the target position 112 and the separation distance L2 between the user and the suspicious person X are considered. ing. Specifically, the guide route candidate R having a larger total distance L1 is less likely to be set as a guide route, and the guide route candidate R having a larger separation distance L2 is more likely to be set as a guide route. Therefore, not only a difficult-to-find guide route but also a guide route that can reach the target position 112 as soon as possible and can be as far as possible from the suspicious person X can be presented to the user. Information can be provided.
By the way, the present invention is not limited to the above-described embodiment, and may be implemented in various different embodiments within the scope of the technical idea described in the claims. Further, the effects described in the embodiments are not limited to this.

  In each of the above embodiments, an infrared sensor is used as the detection unit S, and the position of the person is detected by the detection output of the infrared sensor and the installation position of the infrared sensor (ST2) and is tracked in time (ST3). ). However, the present invention is not limited to this, and an imaging device such as a visible image camera or an infrared camera may be used as the detection unit S to detect the position of a person and track it in time. For example, a wide-angle camera is installed on the ceiling or the like, the position of the person is estimated in ST2 based on the captured image from the wide-angle camera, and the person is tracked in ST3. At this time, for example, a known moving object tracking technique such as a method described in JP 2011-227647 A is used. That is, a moving object is detected from a difference image between a background image and a captured image stored in advance, the position of the moving object in the monitoring space is estimated from the image position of the moving object, and the moving object is positioned in the monitoring image of the preceding and following frames. The moving object is temporally tracked by associating based on the relationship and size. At this time, it is necessary to take a calibration for associating the captured image with the position of the monitoring space in advance.

In each of the above embodiments, the guidance route generation unit 123 calculates the visual field model 113 representing the visual field range of the suspicious person X using the suspicious person position and the space map 111 (ST11 or ST22). At this time, the calculated visual field model 113 is considered to be uniformly weighted in the entire visual field range. That is, in each of the embodiments described above, the degree of awareness D is calculated based on how much the guidance route candidate R (or the user position at the evaluation time when the guidance route candidate R is moved) is included in the visual field model 113. (ST13 or ST25), even if the guidance route candidate R (or the user position at the evaluation time when moving the guidance route candidate R) is included in any range in the visual field model 113, the degree of awareness of the same value Calculated. However, the present invention is not limited thereto, and when calculating the visual field model 113, the range of the visual field model 113 may be divided and weighted, and the degree of awareness D may be calculated in consideration of the weight value. For example, the guidance route generation unit 123 estimates the front direction of the suspicious person X using the detection output of the detection unit S, and the weight range is larger as the range of the visual field model 113 is located in the estimated front direction. It is good also as a structure which allocates. FIG. 16 is a diagram for explaining an example in which the visual field model 113 is weighted. An arrow indicated by a symbol Xd in the figure represents the front direction of the suspicious person X. The front direction Xd is set as the front direction by obtaining the moving direction from the time transition of the suspicious person position. In addition, when an imaging device is used as the detection unit S, the direction in which the front of the face is facing is set as the front direction by using a known face orientation determination technique from the captured image of the imaging device. May be.
And the guidance path | route production | generation means 123 allocates as a weight value larger than another range about the predetermined range of the visual field model 113 which exists in the set ± 15 degree of the front direction Xd. In the example of FIG. 16, a large weight value (for example, “2”) is assigned to the range of the visual field model indicated by reference numeral 113a, and a small weight value (for example, “1”) is assigned to the range of the visual field model indicated by reference numeral 113b. Is assigned. Using the weighted visual field model 113 calculated in this way, the guide route generation unit 123 calculates the degree of awareness D. That is, as the guidance route candidate R (or the user position at the evaluation time when the guidance route candidate R is moved) is included in the range of the visual field model 113a located in the front direction Xd, The degree D is calculated as a large value. Accordingly, since the guide route candidate R that can exist in the visual field range in the front direction of the suspicious person X is evaluated as a dangerous route that is easier to find, providing a safer guide route to the user Can do. In addition to the above, the degree of awareness D is calculated such that the farther the position is from the suspicious person, the harder it is to be found by applying a smaller weight to the range of the visual field model 113 that is farther from the suspicious person X. May be.

  In each of the above embodiments, the space map 111 is stored in the storage unit 11 as two-dimensional coordinate information. However, the present invention is not limited to this, and components (such as walls, floors, pillars, and fixtures) that exist in the monitoring space ( In particular, a component that becomes an obstacle that blocks visible light) may be stored as three-dimensional shape information that is three-dimensionally modeled. For example, 3D shape information created by 3D CAD based on the monitoring space shape information may be used, or 3D shape information obtained by capturing the 3D shape of the monitoring space with a 3D laser scanner or the like may be used. . At this time, the visual field model 113 generated in ST11 or ST22 is generated as a three-dimensional shape. That is, when generating the visual field model 113, the guide route generating unit 123 is a suspicious person position (horizontal position forming a foot position) in the space map 111 that is three-dimensional shape information, and a position having a predetermined height ( For example, the position of 1.6 m) is set as the light source when generating the visual field model 113 as the height position corresponding to the eye height of the suspicious person. And the guidance path | route production | generation means 123 until the light radiated | emitted radially in the predetermined direction (for example, the predetermined range of the front direction of the suspicious person X) from the said light source is interrupted by the structure of the building B expressed as a three-dimensional shape. A three-dimensional visual field model 113 is generated by geometrically calculating the reach. FIG. 17 illustrates a part of the three-dimensional space map 111 and a three-dimensional visual field model 113. As shown in the figure, the range of view until the light emitted from the position approximately coincident with the three-dimensional position of the eye of the suspicious person X is blocked by the structure of the building B represented by the three-dimensional space map 111 is viewed. Calculate as model 113. By calculating the degree of awareness D using the three-dimensional visual field model 113 thus calculated, it is possible to more accurately calculate the hiding of the line of sight due to the structure of the building B, and to reproduce the visual field range of the suspicious person X more faithfully. Therefore, it is possible to calculate a guidance route that is difficult to find from the suspicious person X with higher accuracy.

  In each of the above-described embodiments, when the person determination unit 122 determines whether a person is a suspicious person or a user, the person detection unit 121 determines based on the first appearance position of the person detected and tracked. (ST5). However, the present invention is not limited to this. An important monitoring area is set in advance in the space map 111, a person who exists in the important monitoring area outside the predetermined entry period is regarded as a suspicious person, and other persons are used as users. May be determined. In addition, an imaging device is used as the detection unit S, and the user is face-authenticated by collation with a face image registered in advance using a captured image from the imaging device, and an unknown person who cannot be face-authenticated is determined as a suspicious person. May be.

  In each of the above embodiments, a guidance system is described in which the monitoring space is set indoors in the building B, the target position 112 is set as the entrance D, and a guidance route is generated. However, the present invention is not limited to this, and a guidance route to the target position 112 is generated by setting the monitoring space as a predetermined area “outdoor” and setting the target position 112 as a position where the safety of the user can be ensured (for example, the position of a police station). Thus, a guidance system that safely guides the user may be used. Also in such other embodiments, the suspicious person position and the user position may be detected using the captured image of the imaging device which is the detection unit S installed outdoors, or the mobile terminal 2 equipped with GPS. The terminal position detected by the above may be set as the user position.

  In each of the above embodiments, the operation is simply described using an example in which both the user Y and the suspicious person X are one person. However, the guidance system according to the present invention is applicable even when there are a plurality of suspicious persons and users. That is, even when there are a plurality of suspicious persons and users, the visual field model 113 is calculated for each suspicious person as in the above embodiments. Then, one or a plurality of guide route candidates R are obtained for each user, the degree of awareness D is calculated using the plurality of visual field models 113 calculated for each suspicious person, and guidance is performed for each user using the degree of awareness D. Find a route. Then, the obtained plurality of guidance routes are transmitted to the mobile terminals 2 of all users. Thereby, the user finds a guide route corresponding to his current position from the plurality of guide routes displayed on the map of the display screen 21 of the mobile terminal, and moves safely based on the guide route. be able to. At this time, the guidance device 1 detects the terminal position and the terminal identification number of the portable terminal (for example, the terminal position and the terminal identification number transmitted from the portable terminal) and detects the terminal position and the terminal identification number. , Information on the guide route customized for each user may be transmitted. For example, it is also possible to transmit only the current position and guide route of the user without displaying the current position and guide route of other users. Further, instead of providing the guidance route as image information on the map, the guidance route may be provided as audio information (for example, “Turn left at the next T-junction”).

  In the first embodiment, the guide route candidate R is evaluated using only the degree of awareness D and the guide route is set. However, the present invention is not limited to this, and as in the second embodiment, the total distance L1 and The separation distance L2 may be calculated, and the guidance route candidate R may be evaluated using these values together with the degree of awareness D to set the guidance route.

DESCRIPTION OF SYMBOLS 1 ... Guidance device 11 ... Memory | storage part 12 ... Control part 13 ... Input part 14 ... Communication part 111 ... Spatial map 112 ... Target position 113 ... Visual field model 114. ..Detection signal table 115 ... tracking table 116 ... various parameters 121 ... person detection means 122 ... person determination means 123 ... guide route generation means 2 ... portable terminal 21 ... display Screen 3 ... Network 4 ... Wireless communication device S ... Detector A ... Detection range B ... Building P ... Passage D ... Exit / exit W ... Window X ... Suspicious Y ... User R ... Guidance route candidate

Claims (5)

A detection unit installed in the monitoring space for detecting a person, a portable terminal possessed by the user, and a guide route that is difficult to find from a suspicious person from the current position of the user to the target position is calculated and the portable terminal A guidance system comprising a guidance device for transmitting a guidance route,
The guidance device includes:
A space map showing a structure of the monitoring space and an installation position of the detection unit, and a storage unit that pre-stores a target position on the space map;
Person detecting means for detecting the position of a person existing in the monitoring space using the output of the detection unit and the space map;
Person determining means for determining whether the person existing at the position of the person is a suspicious person or a user;
A range of reach until light projected radially from a suspicious person position in the spatial map in the direction of a predetermined range is obstructed by the structure as a visual field model, and from the user position to the target position using the visual field model And a guide route generating unit that evaluates each one or a plurality of guide route candidates and generates the guide route candidate that is most difficult to find from the suspicious person as the guide route.   The guidance according to claim 1, wherein the guidance route generation unit estimates the front direction of the suspicious person using the output of the detection unit, and evaluates that the guidance route candidate located in the front direction is easier to find. system.   The guidance route generation means estimates the suspicious person position in an evaluation period after the current time using a time transition of the output of the detection unit, and uses the suspicious person position and a spatial map to view the visual field in the evaluation period. A model is estimated, the user position in the evaluation period is estimated for each guidance route candidate, and the ease of finding from a suspicious person in the evaluation period for each guidance route candidate using the visual field model and the user position. The guidance system according to claim 1 or 2, wherein the guidance route candidate that is most difficult to find from a suspicious person is generated as the guidance route. The structure in the spatial map is stored as a three-dimensional shape;
The field-of-view model reaches the suspicious person's horizontal position in the spatial map until the light projected radially from the position in a predetermined direction around the height position corresponding to the eye position is blocked by the structure. The guidance system according to any one of claims 1 to 3, wherein the guidance system has a three-dimensional shape having a range. The guidance route generation means calculates a separation distance between the suspicious person position and the user position, calculates a total distance of a route from the user position to the target position, and the smaller the separation distance, the suspicious person The guidance system according to any one of claims 1 to 4, wherein the guidance system evaluates that it is easy to find from a suspicious person as the total distance increases.

Images